DMMC Course POPULATION GENETICS & SNP ANALYSIS

Durkan Lecture Theatre, Institute of Molecular Medicine, TCD, St James's Hospital

1645-1745 Thursday 29 June 2006.

Plenary Lecture II - Genetic and molecular epidemiology of breast cancer: Promise and pitfalls of new omics approaches
Prof Bob Millikan (North Carolina Center for Genomics and Public Health, University of North Carolina, USA)

A large population-based case-control of breast cancer, the Carolina Breast Cancer Study (CBCS), provides an example of the interface between molecular biology and epidemiology. The study enrolled African American and white women from a defined geographic region of North Carolina (USA), and includes 2015 incident cases of invasive and in-situ breast cancer and 1808 population controls. The CBCS has been used to investigate specific hypothesis related to both the aetiology and progression of breast cancer (1). Results will be presented regarding prevalence of BRCA1 and 2 mutations (2), polymorphisms in DNA repair genes (3-4), mitochondrial DNA (5) and patterns of somatic mutations in P53 (6). Tumour tissue has been used to define specific subtypes of breast cancer using expression profiling, immunohistochemistry and analysis of somatic genetic alterations (7). Different patterns of genetic and environmental exposures were associated with specific breast cancer subtypes in the CBCS. Breast cancer subtypes were correlated with markers of tumour aggressiveness and showed marked differences in patient survival. Haplotype analysis revealed new loci of genetic susceptibility in African Americans and whites (8-9).

The CBCS shows how population-based epidemiology provides important information on the population prevalence of molecular subtypes, how genes and environment interact to cause specific breast cancer subtypes, and the role of population-based studies as validation datasets for investigating determinants of breast cancer survival. Several obstacles have been encountered, including the potential for false positives, population admixture, the need for functional data, and the existence of rare functional alleles within common haplotypes. These problems are likely to be compounded in the future with the advent of whole genome association studies (10).

References

1. Newman, B., et al. (1995). Breast Cancer Research and Treatment. 34, 51-60.

2. Newman, B., et al. (1998). J. Am. Med. Assoc. 279, 915-921.

3. Millikan, R., et al. (2005). Cancer Epidem. Biomarkers Prev. 14, 2326-2334.

4. Mechanic, L., et al. (2006). Carcinogenesis (in press).

5. Canter, J., et al. (2005). Cancer Res. 65, 8028-8033.

6. Conway, K., et al. (2002. Cancer Res. 62, 1987-1995.

7. Carey, L., et al. (2006). J. Amer. Med. Assoc. (in press).

8. Lin, D., et al. (2005). Genetic Epidemiol. 29, 299-312.

9. Pachkowski, B., et al. (2006). Cancer Res. 66: 2876-2877.

10. Millikan, R. (2002). Epidemiology 13, 472-480.